Memristors and neuromorphic networks for high definition bioelectric diagnostics and therapy | The George Washington University

Memristors and neuromorphic networks for high definition bioelectric diagnostics and therapy

Case ID: 019-030-Adam/Efimov

Bioelectric signals within human body control vital human functions.  Abnormal variation in these bioelectric signals would impact human health, e.g. leading to deadly conditions such as cardiac arrythmias.  The present state of bioelectric signal monitoring relies on microsensors and actuators placed on soft substrates which helps in local data collection but requires external computing devices connected by leads to make interpretation which is both time consuming and bulky.

Researchers at George Washington university have developed a novel idea to incorporate data interpretation, decision making, and generation of bioelectric signals locally using a network of computing chiplets. These chiplets can be realized using existing seminconductor technology or can incorporate new types of devices such as memristors (resistive switches) to increase the compactness and energy efficiency. This implementation has the added advantage of incorporating an artificial neural network in an implantable system in a power and area efficient manner.

Applications

  • Real time bio-signal monitoring and data interpretation
  • Possible implementation of neural networks

Advantage

  • Compact size
  • Energy Efficient
  • Real time monitoring
  • Cost effective
  • Reduces time between detection and treatment

 

Diagram

Description automatically generated

Patent Information:

Title App Type Country Patent No. File Date Issued Date Patent Status
Distributed Neuromorphic Computing for High Definition Bioelectric Diagnostics and Therapy US Utility United States 11,701,002 2/3/2020 7/18/2023 Issued

For Information, Contact:

Michael Harpen
Licensing Manager
George Washington University
mharpen@gwu.edu

Inventors:

Igor Efimov
Gina Adam
Keywords: